DE112016002366T5 - PREDICTING USER REQUIREMENTS FOR A BACKGROUND WITH A SPECIFIC CONTEXT - Google Patents

Abstract

A system is described that identifies a task performed by users of the group of computing devices for a particular context based on search histories associated with a group of computing devices for the particular context. The system determines a first degree of likelihood that the task will be performed by the users of the group of computing devices for the particular context, and determines a second degree of likelihood that the second task will be performed by the users of the group of computing devices for the wider context which includes the particular context and at least one other context. In response to determining that the first degree of probability exceeds the second degree of probability by a threshold amount, and that a current context of a particular computing device corresponds to the particular context, the system transmits information to complete the task for the particular context to the particular computing device ,

Description

BACKGROUND WITH PARTICULAR CONTEXT

 When a user is in an unknown situation, the user can use a computing device to obtain information and facts that can assist the user in completing a particular task to achieve a particular goal. Some computing devices require that the user be able to provide sufficient information (eg, search query terms) to guide the computing device to locate the particular information that the user is looking for. Unfortunately, the user may not be aware of which tasks the user needs to complete, much less the information the user should search for in order to navigate successfully through the unknown situation and reach the particular destination. Thus, without prior knowledge of various actions that the user must take, the user may experience stress and waste valuable time and resources to enter the information into a computing device and suspect search terms when the user attempts to obtain information that is necessary. to achieve a specific goal.

SHORT DESCRIPTION OF THE REVELATION

 In one example, the disclosure relates to a method of identifying a task performed by users of the group of computing devices for the particular context by a computer system based on search histories associated with a group of computing devices for a particular context, determining a first one Degree of likelihood by the computer system that the task is performed by the users of the group of computing devices for the particular context and determining a second degree of likelihood by the computer system that the task is performed by the users of the group of computing devices for a wider context which includes the particular context and at least one other context. The method further includes, in response to determining that the first degree of probability exceeds the second degree of probability by a threshold amount, and that a current context of a particular computing device corresponds to the particular context, transmitting information to complete the task for the particular context to the particular computing device, through the computer system.

In another example, the disclosure relates to a computer system including at least one processor and at least one module operable by the at least one processor to identify a task performed by users of the group of computing devices for the particular context based on search histories belonging to a group are assigned by computing devices for a particular context, determining a first degree of likelihood that the task performed by the users of the group of computing devices for the particular context and determining a second degree of likelihood that the task is performed by the Users of the group of computing devices are executed for a wider context that includes the particular context and at least one other context. The at least one module is further operable by the at least one processor to retrieve information in response to determining that the first degree of probability exceeds the second degree of probability by a threshold amount and that a current context of a particular computing device corresponds to the particular context to complete the task for the particular context to the particular computing device.

In another example, the disclosure relates to a computer-readable storage medium containing instructions that, when executed, configure one or more processors of a computer system to identify a task performed by users of the group of computing devices for the particular context based on search histories associated with a group of computing devices for a given context, determining a first degree of likelihood that the task is performed by the users of the group of computing devices for the particular context, and determining a second degree of likelihood that the task is performed by the users of the group of computing devices for a broader context that includes the particular context and at least one other context. The instructions, when executed, further configure the one or more processors of the computer system to respond to determining the first degree of probability exceeds the second degree of probability by a threshold amount and that a current context of a particular computing device conforms to the particular context is equivalent to transferring information for completing the task for the particular context to the particular computing device.

The details of one or more examples are set forth in the accompanying drawings and the description below. Other features, objects and benefits of Disclosure will be apparent from the description and drawings, and from the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

1 FIG. 3 is a conceptual diagram illustrating an exemplary system for predicting the needs of a user of a computing device and providing information to meet the needs in accordance with one or more aspects of the present disclosure.

2 FIG. 3 is a block diagram illustrating an exemplary computer system configured to predict the needs of a user of a computing device, and to provide information to meet the needs, in accordance with one or more aspects of the present disclosure.

3 FIG. 10 is an additional conceptual diagram illustrating an exemplary system for predicting the needs of a user of a computing device and providing information to meet the needs in accordance with one or more aspects of the present disclosure.

4A and 4B 13 are flowcharts illustrating example operations performed by an exemplary computer system configured to predict the needs of a user of a computing device, and to provide information to meet the needs, in accordance with one or more aspects of the present disclosure.

DETAILED DESCRIPTION

 In general, techniques of this disclosure may enable a computer system to learn and predict the needs of a user of a computing device for a particular context based on previous actions of other users of other computing devices for the particular context. Based on contextual information of the user (e.g., location, time, user interests, and the like), the system may derive a task or action that the user may need to perform and automatically provide information that users of other computing devices need in similar contexts, to complete the task or perform the action. The computer system may classify the information that the computer system provides for completing a task or performing an action by determining degrees of usefulness of the information for different contexts. The computer system may provide the information for completing a task or performing an action if the degree of usefulness for the particular context is sufficiently greater than the degree of usefulness for performing the action of other contexts.

Throughout this disclosure, examples are described in which a computing device and / or computer system only analyzes information associated with a computing device and a user of a computing device (eg, context, locations, speeds, search queries, etc.) when the computing device has a Receive permission from the user of the computing device to analyze the information. Thus, in situations described below, before a computer or computer system is allowed to collect or use information associated with a user, the user may be given an opportunity to make an input to control programs or functions of the computing device and / or the computer system To collect and use user information (such as information about a user's current location, current speed, etc.), or to specify whether and / or how the device and / or system can receive content that may be of use to the user. are relevant. In addition, certain data may be treated in one or more ways before being stored or used by the computing device and / or computer system so that personally identifiable information is removed. For example, a user's identity may be treated so that no information for identifying people via the user may be determined, or a user's geographic location when location information is received (for example, a city, postal code, or country level) so that a specific location of a user can not be determined. Therefore, the user may have control over how the information about the user is collected and used by the computing device and computer system.

1 is a conceptual diagram, the system 100 as an exemplary system for predicting the needs of a user of a computing device 110A and provides information to meet the needs in accordance with one or more aspects of the present disclosure. system 100 includes an information server system ("ISS") 160 that over the network 130 in communication with the computing devices 110A - 110N (summarizing "Computer Devices 110 ") stands.

network 130 represents any public or private communications network, such as a cellular network, Wi-Fi, and / or other network types for transferring data between computer systems, servers, and Computer equipment. network 130 may include one or more network hubs, network switches, network routers, or any other network equipment that is operatively interconnected, thereby facilitating the exchange of information between the ISS 160 and the computer devices 110 provided. computer equipment 110 and ISS 160 Data can be sent over any network using any suitable communication techniques 130 send and receive.

ISS 160 and computer equipment 110 can each be operational with network 130 be connected using appropriate network connections, and computing devices 10N can be operational with network 30B be connected using a different network connection. The connections, the computing devices 110 and ISS 160 with network 130 may be Ethernet, ATM or other types of network connections, and such connections may be wireless and / or wired connections.

computer equipment 110 represent individual mobile devices, such as A mobile phone, a tablet computer, a laptop computer, a computerized watch, computerized glasses, computerized gloves, or other type of portable computing device. Additional examples of computing devices 110 include personal digital assistants (PDAs), portable gaming systems, media players, e-book readers, mobile television platforms, automotive navigation and entertainment systems, or any other types of mobile, portable, and non-portable computing devices configured to share information over a network , such as network 130 , to recieve.

Each of the computer devices 110 includes a corresponding user interface device 112A - 112N (summarizing "UIDs 112 ") And a corresponding user interface module 120A - 120N (summarizing "UI modules 120 "). In addition, each of the computer devices includes 110 a corresponding query module 122A - 122N (in summary, "query modules 122 "). modules 120 - 122 may perform described operations using software, hardware, firmware, or a combination of software, hardware, and firmware included in the respective computing devices 110 resident

and / or running on it. computer equipment 110 can each modules 120 - 122 with multiple processors or multiple devices. computer equipment 110 can each modules 120 - 122 as virtual machines running on underlying hardware. modules 120 - 122 may be executed as one or more services of an operating system or a computer platform. modules 120 - 122 may be executed as one or more executable programs at an application level of a computer platform.

computer equipment 110B - 110N Form a group of computing devices, each of which is the computer devices 110B - 110N associated with searches for information. In some examples, computing device is 110A in the group with computer devices 110B - 110N includes, and a user, the computing device 110A It can also search for information from computer device 110A To run. In other examples, the group of computing devices includes 110B - 110N computing device 110A out.

UIDs 112 the computer devices 110 can as appropriate input and / or output devices for computer equipment 110 act. UIDs 112 can be implemented using different technologies. UIDs 112 For example, using presence-sensitive input screens such as Resistive touchscreens, Surface Acoustic Wave (SAW) touchscreens, capacitive touchscreens, projective-capacitive touchscreens, pressure-sensitive screens, Acoustic Pulse Recognition (APR) touchscreens, or other presence-sensitive touchscreens Display technology act as an input device. In addition, UIDs 112 Include microphone technologies, infrared sensor technologies, or other input device technology for use in receiving user input.

UIDs 112 can be measured using any one or more display devices, such. LCD (Liquid Crystal Display), dot matrix, LED ("Light Emitting Diode"), OLED (Organic Light Emitting Diode), E-paper, or similar Monochrome or color displays that provide visible information to a user of computing devices 110 can act as output devices (eg, display). In addition, UIDs 112 Speaker technologies, haptic feedback technologies, or other output device technology for use in delivering information to a user.

UIDs 112 may each include presence-sensitive displays that include tactile input from a user of the respective computing devices 110 can receive. UIDs 112 can receive clues to tactile inputs by typing Detect one or more gestures from a user (eg, the user using a finger or stylus to identify one or more locations of UIDs 112 touches or points to it). UIDs 112 can present issues to a user, e.g. On respective presence-sensitive displays. UIDs 112 The output can be configured as respective graphical user interfaces (eg user interface 114 ), that of computer equipment 110 provided functionality can be assigned. For example, UIDs 112 different user interfaces (eg user interface 114 ), which relate to search functions used by query modules 122 or other features of computer platforms, operating systems, applications and / or services operating on computing devices 110 be executed or accessible (e.g., electronic news applications, internet browser applications, mobile or desktop operating systems, etc.).

UI modules 120 manage user interactions with respective UIDs 112 and other components of the computing devices 110 , UI modules 120 can cause UIDs 112 output respective user interfaces, such as the user interface 114 (or other example user interfaces) to indicate when users of the computing devices 110 View issues and / or UIDs 112 Provide input. UI modules 120 and UIDs 112 can obtain one or more indications of user input at different times when users interact with the graphical user interfaces and when the users and computing devices 110 located in different locations. UI modules 120 and UIDs 112 can UIDs 112 interpret recognized inputs and those on UIDs 112 redirected information about the inputs to one or more linked platforms, operating systems, applications, and / or services that reside on the computing devices 110 be executed to z. B. Computer equipment 110 to cause functions to be performed.

UI modules 120 may receive information and instructions from one or more linked platforms, operating systems, applications, and / or services that reside on the computing devices 110 and / or one or more remote computer systems, e.g. The ISS 160 , In addition, UI modules 120 act as intermediates between the one or more linked platforms, operating systems, applications and / or services running on computing devices 110 and the different output devices of the computing devices 110 (eg, speakers, LED displays, audio or electrostatic haptic output devices, etc.) to output (eg, a graphic, a flash of light, a sound, a haptic response, etc.) to computer devices 110 to produce.

In the example of 1 is user interface 114 a graphical user interface associated with the demand forecasting service provided by ISS 160 is provided, and on the computing device 110A accesses. As described in detail below, user interface includes 114 graphical information (eg, text) that represents information, according to ISS 160 Prediction stating that a user of computer device 110A at the current time may need to perform a task. user interface 14 may include various other types of graphical information, such as: B. visual representations of predicted information that a user of the computing device 110A may be needed to perform a task for a current context. UI module 120A can the UID 112A to output a user interface 114 based on data causing the UI module 120A over the network 130 from the ISS 160 receives. The UI module 120A can display graphical information (eg text data, image data, etc.) to present the user interface 114 as input from ISS 160 along with instructions from ISS 160 to present the graphical information in UI 114 to UID 112 receive.

query module 122 perform search-related functions for computing devices 110 out. query module 122 can search queries (eg strings) over network 130 to ISS 160 send and in response results of those of ISS 160 searches performed based on the queries received. query module 122 can provide pointers to the search queries of UI modules 120 obtained when users of computer devices 110 Entries to UIDs 112 provide. query module 122 can search results provided by ISS 160 received at UI modules 120 spend to B. UI modules 120 to get the search results as part of UI UI modules 120 that at UIDs 112 present, present.

ISS 160 represents any suitable remote computer system, such as a computer. For example, one or more desktops, laptops, mainframes, servers, cloud computing systems, and so on are able to transfer information to and from a network, such as a network. Eg network 130 to send and receive. ISS 160 hosts (or at least provides access to) a search system for searching for information and requires a needs prediction system for predicting the needs of users of computing devices 110 for specific contexts.

computer equipment 110 can with ISS 160 over network 130 Communicate to that of ISS 160 provided search system and needs prediction system. In some examples, ISS represents 160 a cloud computing system that provides search and forecasting services through the network 130 for one or more computing devices 110 accessing the search and forecasting services through cloud access provided by ISS 160 provided.

In the example of 1 includes ISS 160 context module 162 , Needs prediction module 164 and search engine 166 , modules 162 - 166 can perform described operations using software, hardware, firmware, or a combination of hardware, software, and firmware available from ISS 160 is resident and / or executed. ISS 160 can modules 162 - 166 with multiple processors or multiple devices. ISS 160 can modules 162 - 166 as virtual machines running on underlying hardware. modules 162 - 166 may be executed as one or more services of an operating system or a computer platform. modules 162 - 166 may be executed as one or more executable programs at an application level of a computer platform.

context module 162 Collects contextual information related to computing devices 110 and provides the information for modules 164 and 166 ready and make determinations based on contextual information. For example, context module 162 computing device 110A get assigned location information and determine if the current location of computing device 110A with a previous location from a location history of computing device 110A matches. context module 162 Can maintain contextual histories of any of the computing devices 110 are assigned and determine whether a respective current context, the one or more of the computing devices 110 associated with a previous context found in a particular context history. For example, context module 162 as part of a computer device 110A associated context history maintain a location history that tracks where computing device 110A was positioned on a particular day or time.

context module 162 can continuously update the current context and context history, the computing device 110 are assigned when context module 162 receive new and updated information. For example, context module 162 from sensors of the computer device 110A receive collected motion information, and use the motion information to update the current location, the context module 162 for computer device 110A is maintained.

Location and motion information is just a few types of context information, the context information module 162 for each of the computing devices 110 can sustain. As used throughout the disclosure, the term "context information" is used to describe imaginable information provided by a computer system and / or computing device, such as ISS 160 and computer devices 110 be used to determine one or more environmental or behavioral characteristics associated with computing devices and / or users of computing devices.

For example, context information includes past, current, and future physical locations, degrees of movement, weather and traffic conditions, travel patterns, and the like. In some examples, context information may include sensor information provided by one or more sensors (eg, gyroscopes, acceleration sensors, proximity sensors, etc.) of computing devices 110 Radio transmission information transmitted by one or more communication units and / or radios (eg Global Positioning System (GPS), cellular, Wi-Fi, etc.) from computing devices 110 information obtained from one or more input devices (eg cameras, microphones, keyboards, touchpads, mice, UID 112 etc.) of computer devices 110 and network / device ID information (e.g., a network name, an IP address of a device, etc.) of the computing devices 110 to be obtained.

In addition, context information may include user topics of interest (e.g., a user's preferred "things" typically maintained as a user interaction chart or any other type of data structure), contact information associated with users (eg, a user's personal contact information as well Information about the user's friends, employees, social media, families, etc.), search histories, location histories, long and short-term tasks, calendar information, application usage histories, purchasing histories, favorites, bookmarks, and other information that the computing devices provide 110 and ISS 160 about a user of the computing devices 110 can collect.

Furthermore, context information may include information about the operating state of a computing device. For example, an application that runs at a specific time or location is one Example of information about the operating state of a computer device. Other examples of contextual information based on the operating state of a computing device include, but are not limited to, positions of switches, battery charge states, whether a device is plugged into a wall outlet or otherwise operatively connected to another device and / or machine, user authentication information (e.g. which user), whether a device is operated in "airplane" mode, in standby mode, in full power mode, the operating state of radios, communication units, input devices and output devices, etc.

context module 162 can provide the context information and make provisions about the context information, the context module 162 maintains, in favor of other modules 164 and 166 and the computer devices 110 , For example, the context module 162 upon a request of the needs prediction module 164 from ISS 160 respond to contextual information, the computing devices 110 assigned for transmission to need prediction module 164 , from context module 162 maintained data specifying the context information (eg, respective locations) associated with one or more of the computing devices 110 assigned.

It should be understood that although throughout this disclosure examples are described in which a computing device and / or computer system analyzes contextual information, including location history, contextual histories, and / or search histories associated with a computing device and a user of a computing device, the computing devices described herein and Computer systems can analyze only the context information when the computer device and / or computer system receives the explicit permission from the user of the computer device to analyze the information. Thus, in situations discussed herein, before a computing device or computer system is allowed to collect or use information associated with a user, the user may be provided with an opportunity to provide input to control programs or functions of the computing device and / or computer system Collect and use user information (such as current location information, user's current speed, etc.), or to specify whether and / or how the device and / or system can receive content that may be of use to the user. are relevant. In addition, certain data may be treated in one or more ways before being stored or used by the computing device and / or computer system so that person-identifiable information is removed. For example, a user's identity may be treated so that a person-identifiable information about the user can be determined, or a user's geographic location can be generalized when location information is received (e.g., a city, postal code, or country level). so that a specific location of a user can not be determined. Therefore, the user may have control over how the information about the user is collected and used by the computing device and computer system.

search module 166 performs searches for information (such as those available through the Internet) based on search queries, the search engine 166 of computer devices 110 and / or modules 162 and or 164 receives. For example, search engine 166 Do an Internet search of a string, the search engine 166 from the query module 122B receives when a user of the computer device 110B with the computer device 110B interacts to locate an online image of an airport terminal card. In response to the search, search engine 166 get a graphic image of the airport terminal card and the graphic image back to computer device 110B output.

The search engine 166 can scan one or more users of the computer device 110 maintained. The through search modules 166 maintained search histories can be part of the context module 162 maintained contextual histories, or be separate from it. The through search modules 166 generated search histories can be sortable and searchable, so search engine at a given time 166 the search histories of one or more of the computing devices 110 can analyze to determine what types of queries users of the computing devices 110 examined. search module 166 can modules 162 and 164 Provide access to the search histories and / or analyze the search histories and information from the search histories in the name of modules 162 and 164 output.

Need prediction module 164 learns and says the needs of users of computing devices 110 for different contexts previously, and can automatically provide information for computer devices 110 that users need from other computing devices in similar contexts to complete a task or perform an action. For example, the needs prediction module 164 in view of certain previous actions (eg, searching for particular queries) made by a user of a computing device and / or other users of other computing devices, a task or a task Infer action that the user may need to perform for a particular context.

Based on the information, the need prediction module 164 of context module 162 about current contexts of computer devices 110 can needs-prediction module 164 determine when a current context corresponds to a previous context within which users of computing devices 110 the task or action previously performed. Need prediction module 164 can the search engine 166 request to perform a search for information that can assist a user in completing the derived action or task. The need prediction module 164 can the information, the search engine 166 to complete the task or perform the action, find one or more computing devices 110 for presentation to a user (eg as a user interface 114 at one of the UIDs 112 ) provide.

A user "need" as described herein may be defined as a semantically grouped pattern of user actions in a given context based on observations by a system of user actions (eg, in action logs such as search histories, purchase histories, application usage histories and other contextual histories). It should be understood that this is only an example definition for simplifying the discussion of the need prediction module 164 used model is. Therefore, because of this example definition, the tasks or actions may not necessarily be needs; rather, observed actions or performed tasks may be a proxy, the needs prediction module 164 can use to the more general needs of users of computer equipment 110 derive.

Need prediction module 164 With respect to the additional FIGS. described in more detail. In summary, needs prediction module 164 perform one or more machine learning techniques to meet the needs of users of computing devices 110 to learn for different contexts. In other words, model needs prediction module 164 User needs, since need prediction module 164 possibly incapable of meeting the needs of users of computing devices 110 to learn directly. Need prediction module 164 can both context information from context module 162 as well as search history information from search engine 166 and provide this information as inputs to the model for training the model to learn the needs of users for particular contexts. Once trained, the model may receive a particular context as an input, and in response, issue a hint (eg, data) indicative of one or more tasks, actions, or other observable actions that were previously performed by other users for the purpose specific context.

According to the techniques of this disclosure, ISS 160 based at least in part on search histories (eg, in addition to purchase histories, application usage, etc.) associated with a group of computing devices 110 For a given context, identify a task performed by users of the group of computing devices 110 was executed for the particular context. For example, a user of the computing device may 110A Visit another country for the first time and take a flight to an airport in the other country. Upon arrival, the computer device can 110A a connection via network 130 with ISS 160 build up.

context module 162 can be a current context of the computing device 110A determine and the current context with need prediction module 164 share. Need prediction module 164 can change the current context of the computing device 110A into a model that contains the needs prediction module 164 used to derive one or more actions, typically by other users of computing devices 110 in the face of similar contexts. Need prediction module 164 Using the model, it can identify one or more tasks that users typically perform when visiting the airport of the other country.

In some examples, an example of a derived task includes navigating to a particular location from a current location of the computing device 110A , For example, the model of the needs prediction module 164 Output data that suggests that users typically search for taxis from the airport in the other country. Or, in some examples, the model may output data that suggests that users are typically performing (eg, opening) a ground transportation related application from their respective one of the devices 110 trigger to order a taxi.

In any case, need prediction module 164 Based on the information received from the model, users of the computing devices determine 110 typically need a taxi upon arrival at the airport, and may therefore deduce that the user of the computer device 110A could need a taxi. In some examples, needs prediction module 164 easy search engine 166 request a search based on the tasks identified by the model or Perform actions. Need prediction module 164 can then automatically send the search results (or the information derived from the search results) to the computing device 110A spend so that UI module 120A alert the user that he or she may need to call a taxi.

In other examples, needs prediction module 164 perform some additional analysis of the information received from the needs model before sending information to the computing device 110A sends. Simply put, needs prediction module 164 determining whether a degree of likelihood (eg, a probability, a count, etc.) that the task is performed for the particular context sufficiently compensates for a degree of likelihood that the task will be performed for a wider context. In this way, needs prediction module 164 only output information if the task that the user may need to perform is out of the ordinary or otherwise unique to a particular situation.

For example, the needs model of the needs prediction module 164 Determine that users typically use an e-mail application from computing devices 110 open when they are at a specific airport. The needs model of the needs prediction module 164 however, may determine that users typically share the same e-mail application from computing devices 110 open for other contexts, including the airport and a large geographical area surrounding the particular airport. Need prediction module 164 may determine that the probability of opening the e-mail application upon arrival at the airport in the other country is similar to the degree of likelihood of opening the e-mail application outside the airport in the other country. Therefore, need prediction module 164 determine that the user of the computer device 110A It is not necessary to be aware of receiving an alert to check his or her email, as the "need" may not be of sufficient interest to interrupt or otherwise proactively alert the user.

In order to determine whether information for a need to accomplish a particular task is of sufficient interest to interrupt a user, the needs model of the needs prediction module may 164 a first degree of probability that the task by the users of the group of computing devices 110 for the particular context is executed, compare with a second degree of probability that the task by the users of the group of computing devices 110 for a broader context that includes the particular context and at least one other context. For example, the needs prediction module 164 determine that seventy-five percent of users of computing devices 110 Searching for ground transportation means when the users are at the airport in the other country, and that only twenty percent of users of the computing devices 110 perform the same ground transportation search when the users are anywhere within the other country, including when the users are at the airport.

The needs model of the needs prediction module 164 may compare the difference between the two probabilities with a threshold amount (eg, ten percent, twenty percent, etc.) to determine if the need for information for the particular context sufficiently balances the need for similar information in the broader context. For example, the threshold amount may be twenty percent, or in other words, the need model of the need prediction module 164 require that the need for information in the particular context outweighs the need for information in the wider context by at least twenty percent.

In response to determining that the first degree of probability exceeds the second degree of probability by a threshold amount, and that a current context of the computing device 110A may correspond to the particular context may need prediction module 164 Information about completing the task for the particular context on the computing device 110A transfer. For example, needs prediction module 164 after determining that probability that a user of the computing device 110A Soil transport at the airport in the other country will need to be sufficiently larger than the likelihood that the user will need ground transportation anywhere in the country, search engine 166 to carry out a search for useful information in connection with the use of ground transportation from the airport in the other country.

In some examples, the information to complete the task for the particular context, the ISS 160 computing device 110A based on at least one search result based on the search histories of the group of computing devices 110 assigned. For example, the computer device 110A Some or all of the information sent to the previous user by performing a search in in a particular context. In this way, ISS 160 the information that the user is likely to search anyway, without requiring that the user with his or her computer device 110A interacts to perform the search.

In some examples, ISS 160 on transferring the information to computer device 110A to complete the task for the particular context in response to determining, dispense with the first degree of probability not exceeding the second degree of probability by the threshold amount. For example, the needs prediction module 164 after determining that the probability that the user of the computing device 110A Soil transport at the airport in the other country will need not be sufficiently greater than the likelihood that the user will need ground transportation anywhere in the country to forego search module 166 to initiate a search for useful information in connection with the use of ground transportation from the airport in the other country. In other words, ISS 160 Information to meet a predicted need may not be ready if the need is not unique to a particular situation, but instead is generally needed for many or all situations.

In any case, search engine 166 assuming that the first degree of probability exceeds the second degree of probability by the threshold amount, and needs prediction module 164 search module 166 to perform a search for information associated with a need to return information about the need. In some examples, the information may include one or more suggestions for traveling from a current location to a particular location. In some examples, the information may include the location of the taxi depot at the airport, whether tipping is common in the other country, and other useful information about the use of ground transportation from the airport in the other country. For example, search engine 166 determine that only local currency (eg no foreign currency or credit cards) will be accepted by ground transport operators from the airport. Need prediction module 164 can send information to computer device 110A Issue the instructions for presenting the UI 114 to UID 112A includes to the search engine 166 information received.

After receiving the information from ISS 160 can UI module 120A UID 112A induce user interface 114 to display and, in some examples, may be computing device 110A to issue an acoustic alarm as well as to alert the user to useful information for meeting his or her needs in the particular context on UID 112A Are available. In this example, the user of the computing device 110A based on the ISS 160 The information received informs him that he should visit a currency exchange counter before boarding a taxi.

Accordingly, if a user of a computing device receiving information from an example computer system performing the techniques described herein is in an unknown situation, it may be necessary for the user to perform searches for information or otherwise be inputs to or Provide their computing device to obtain information and facts that can assist the user in completing a particular task to achieve a particular goal. Instead, the example system may automatically derive the user's needs from the particular context of the situation, and automatically provide information useful to the computing device that meets the needs of the user and may assist the user in achieving the task or goal.

In this way, unlike other computing devices and systems that require the user to type inputs (eg, search queries), and further, to be able to provide sufficient information (eg, correct search query terms). To provide the other computing devices and systems with locating the particular information that the user is looking for, the example system automatically provides the needed information. Even if the user is unaware of the tasks that the user may need to complete or the information that the user should search to successfully navigate through the unknown situation and reach the particular destination, the exemplary system will still be automatic provide the information you need

In addition, unlike other computing devices and systems, the exemplary system may provide needed information that has a particular "wow!" Effect. In other words, because the output of information for a particular context is dependent on whether a first degree of probability of a task performed by users of the group of computing devices for a particular context may have a second degree of probability of the task by the users of the group is performed by computing devices for a wider context and at least one context exceeds a threshold amount, the example computer system is more likely to provide information that a user does not already know and / or from which he has no way of knowing that he or she would need them for a particular context. The information provided by the example system may include some of the more covert facts or other nuanced information that is very specific to a particular situation of a user (eg, time, location, etc.) that would otherwise be known only to very experienced users who were in the specific situation before.

Consequently, the user does not need to have prior knowledge of the different actions that the user may need to perform before experiencing the unknown situation. The user may experience less stress and spend less time providing input to a computing device to search for information as the user attempts to obtain information necessary to achieve a particular goal. With less direct user interaction (eg, input) by a user, the exemplary system can save energy and use less battery power compared to other systems that merely provide a user with the ability to search for information.

2 is a block diagram, the Information Server System (ISS) 260 as an exemplary computer system useful for predicting the needs of a user of a computing device, such as a computer. B. one of the computer devices 110 from 1 and configured to provide information to meet the needs in accordance with one or more aspects of the present disclosure. ISS 260 is a more detailed example of the ISS 160 from 1 and is down within the context of system 100 from 1 described. 2 only illustrates a specific example of the ISS 260 and many other examples of the ISS 260 can be used in other instances and include a subset of the components that are included in the example ISS 260 or contain additional components that are included in 2 are not shown.

ISS 260 Represents computer devices 110 a line ready through which a computer device, such. B. Computer equipment 110A , Searches for information related to search queries, and in some instances can automatically receive information from which the ISS 260 predicts that it meets the needs of users of computing devices 110 for a particular context. As in the example of 2 shown, includes ISS 260 one or more processors 270 , one or more communication units 272 and one or more storage devices 274 , storage devices 274 of the ISS 260 include context module 262 , Needs prediction module 264 and search engine 266 , Within the needs prediction module 264 include storage devices 74 training module 268 , modules 262 - 266 contain at least the same, if not more, abilities than each module 162 - 166 from 1 ,

storage devices 274 of the ISS 260 also include search history data storage 270A , Context History Datastore 270B and task and requirement rule data storage 270C (summarizing "Datastores 270 "). communication channels 276 can work with any of the components 270 . 272 and 274 be connected for communication between the components (physical, communicative and / or operative). In some examples, communication channels may be 276 include a system bus, a network connection, a cross-process communication data structure, or any other method of communicating data.

One or more communication units 272 of the ISS 260 can be used with external computing devices, such as B. computer devices 110 from 1 communicate by transmitting and / or receiving network signals on one or more networks, such as, e.g. Eg network 130 from 1 , For example, ISS 260 communication unit 272 for transmitting and / or receiving radio signals via network 130 for exchanging information with computer devices 110 use. Examples of communication unit 272 They may include a network interface card (eg, an Ethernet card), an optical transceiver, a radio frequency transceiver, a GPS receiver, or any other type of device that can send and / or receive information. Other examples of communication units 272 may include shortwave radios, cellular data radios, wireless Ethernet network radios, and Universal Serial Bus (USB) interfaces.

One or more storage devices 274 within the ISS 260 can provide information for processing during the operation of the ISS 260 save (for example, the ISS 206 Save data on the modules 262 . 264 . 266 and 268 during the execution of the ISS 260 can access). In some examples, the storage devices 274 a temporary memory, which means that the main purpose of the storage devices 274 not the long-term storage is. The storage devices 274 on the ISS 260 can be configured as volatile memory for short-term storage of information, and therefore can not retain stored content when turned off. Examples of volatile memory include main access memory (RAM), dynamic RAM (DRAM), static random access memory (SRAM), and other forms of volatile memory known in the art.

The storage devices 274 also include one or more computer-readable storage media in some examples. The storage devices 274 can store larger amounts of information than volatile memory. The storage devices 274 may also be configured for long-term storage of information as nonvolatile storage space and for maintaining information after power on / off cycles. Examples of nonvolatile memories include magnetic hard disks, optical disks, floppy disks, flash memories or forms of electrically programmable memories (EPROM) or electrically rewritable and programmable (EEPROM) memories. The storage devices 274 can program instructions and / or data in connection with the modules 262 . 264 . 266 and 268 to save.

One or more processors 270 can implement functionality and / or instructions within the ISS 260 To run. For example, processors can 270 on the ISS 260 Receive and execute instructions issued by the storage devices 274 stored are the functionalities of modules 262 . 264 . 266 and 268 To run. These through the processors 270 Instructions executed may be the ISS 260 to do so during program execution in the storage devices 274 Store information. The processors 270 can get instructions from the modules 262 . 264 . 266 and 268 to learn and predict the needs of users of computing devices for different contexts in light of the previous actions of other users of other computing devices for the same contexts, and to automatically provide information relevant to the predicted needs of a user of a computing device for a particular context , That means the modules 262 . 264 . 266 and 268 from the processors 270 can be used to numerous actions, or functions of the ISS 270 as described in this application.

The data storage 270 represent any suitable storage medium for storing information associated with search histories (e.g., search terms, synonyms, related search terms, etc.), contextual histories (e.g., locations, times of day, device state information, user information, etc.) and rules (e.g. a machine learning system) for predicting the needs of users of computing devices for different contexts. The in the data stores 270 stored information may be searchable and / or categorized such that one or more modules 262 - 268 Input request information from one or more of the data stores 270 and in response to input to the data stores 270 receive stored information.

The search history data store 270A can primarily through the search engine 266 can be part of the context history datastore 270B , or be separate from it, generally by the context module 262 is maintained. The search history data store 270A may include one or more searchable databases or data structures that organize the different types of information searched by individual computing devices, such as: B. the computer devices 110 from 1 , For example, the search history data store 270A Include lines dedicated to individual or groups of computing devices, and within each line the search history data store 270A Information related to searches include the search engine 266 in favor of the individual, or groups of computing devices engineered. Examples of types of in the data store 270A Information stored includes search terms or requests, times of day and / or locations of the computing devices associated with the search terms, counts associated with the search terms indicating how frequently or how often searches for the search term occurred, an indication of whether a search of a search term was successful (eg, whether the search results directly in a subsequent search, which implies that the search was unsuccessful), and any and all other information related to searches performed by the computing devices.

As noted above, the context history data store may be 270B some or all of search history data storage 270A include. Context history data storage 270B includes organized and searchable context history information associated with individual, or groups of computing devices, such as computer tools. B. Computer equipment 110 , These types of contextual information stored in the context history datastore 270B may include location information, time of day information, sensor information (eg from computer devices 110 receive user information information, information about an operating state of the device, application execution information (eg, which, and when the application was executed), and all other information that is the need prediction module 264 may need to predict the needs of users.

The task and need rules data store 270C includes one or more previously developed rules on which the need prediction module 264 to predict a task or action that is likely to be performed by a user of a computing device for a current context, as well as information that the user may need to complete the task. For example, the data store 270C Rules of a machine learning or artificial intelligence system of the needs prediction module 264 to save. The system for machine learning or artificial intelligence of the needs prediction module 264 can be based on the rules of the data store 270C access to tasks and needs derived from the users of the computing devices 110 are assigned for a particular context.

In some examples, the needs prediction module 264 a current context of a computing device as input to the data store 270C and receive as output information associated with a task or action that a user of the computing device may need to execute in the current context and context (which includes the current context). In some examples, the rules of datastore 270C output a degree of probability (eg, a probability, a count, etc.) associated with the task or action for the current context, and a similar degree of probability associated with the task or action for the wider context. And in some examples, the need prediction module 264 the predicted task as input to the data store 270C and receive as output information related to the types of information the user may need in the current context to complete the predicted task.

The training module 268 of the needs prediction module 264 can be in the data store 270C Generate stored rules to predict and determine the degrees of probability associated with the tasks performed by users for particular contexts. For example, a system for machine learning or for artificial intelligence of the training module 268 through the context module 262 received, and in the data memory 270B analyze stored context information and correlate with those in the data store 270A identify stored search information. Types of through the training module 268 The machine learning systems used include Bayesian networks, neural networks, and other types of artificial intelligence models. For example, the training module 268 a table with one line for each computing device 110 or a group of the computing device 110 which share similar content and similar queries running in similar contexts. Based on the correlations between context information and search information provided to the computing devices 110 can be assigned to the training module 268 To develop rules for predicting future searches by computer devices 110 for certain contexts.

For example, the training module 268 determine that there is a correlation between search queries associated with terminal maps for a particular airport when a section of the computing device 110 is positioned at or near the designated airport. The training module 268 may develop a rule that produces a high degree of likelihood that a computing device will perform a search for the terminal map of the airport when the device is positioned at that particular airport.

As another example, the training module 268 determine that there is a correlation between search queries associated with weather during the morning hours of a day when a section of computing devices 110 is positioned at or near the respective home location, the section of computer equipment 110 assigned. The training module 268 For example, in turn, may develop a rule that produces a moderate degree of likelihood that a computing device will seek a search for the weather during the morning hours of future days when the device is positioned at a home location. The training module 268 can produce an infinite number of rules to predict the types of future actions based on past observed actions, the users of computing devices 110 for a particular context.

The training module 268 For example, probabilities P of users having different needs N may learn in different ways according to contexts C. In some examples, the training module may 268 Determine P (N | C) using multinomials or counts. For example, the training module 268 determine how many times computer devices 110 Searches for an airport terminal card while in the airport were compared to the total number of searches made from within the airport to determine a ratio of terminal card searches to total searches.

In some examples, the training module may 268 Determine P (N | C) using Bayesian rule. For example, that can training module 268 using the Bayesian rule an equation such. Eg equation GL. use l; P (N | C) is proportional to P (N | C) .P (N) GL. 1

In GL. 1, P (N) is the probability that the user has a need N, and P (C | N) is the probability that the user in the context would be C if they have the need N. In other words, the later probability for a given need that a context be proportional to the probability of the context given the need times of the prior probability of the need.

The training module 268 Further, aggregation techniques and clustering techniques may be performed on the context information and search histories to determine improved rules for predicting a user's needs. For example, the training module 268 aggregate the calculated probabilities for specific contexts to determine probabilities for more general context, since the training module 268 P (N | C) for each specific context that is determined by the context module 262 is determined. For example, the training module 268 determine a probability of a particular need for multiple individual airports, and then aggregate the respective probabilities for each of the different airports into a single probability for all the airports.

The training module 268 can perform clustering techniques to match similar actions with similar needs provided by the search engine 266 were observed. For example, the training module 268 determine that if two different queries resulted in the same search results, the two queries must be similar and therefore treated as similar, if not as the same action or task, for which similar needs are required.

3 is an additional conceptual diagram representing the system 300 , including Information Server System (ISS) 360 as an exemplary system for predicting the needs of a user of a computing device, such as a computer. B. one or more computer devices 110 from 1 , and for providing information to meet the needs, in accordance with one or more aspects of the present disclosure. The ISS 360 is an additional example of the ISS 160 from 1 and the ISS 260 from 2 , and is below in the context of the system 100 from 1 and the ISS 260 from 2 described.

The system 300 includes the ISS 360 that over the network 330 in communication with the computing devices 310A - 310N (summarizing "Computer Devices 310 ") stands. The ISS 360 includes the needs prediction module 364 , the table 370 can produce to match observed actions or tasks performed by the computing devices 310 with respective contexts (eg locations). Based on the in table 370 stored information may be the need prediction module 364 determine when one of the computer devices 310 Information may need to perform a predicted task or action, by need prediction module 364 has been identified as possibly to seize.

3 represents the computing devices 310A . 310B . 310D and 310E at the site 304 which in this case is San Francisco Airport (SFO). The surrounding location 304 is location 302 which in this example represents the ten-square-kilometer area of the state of California surrounding SFO. In other words, the location 304 represent a particular context associated with a first geographic area and the location 302 may represent a broader context associated with a second geographic area that includes the first geographic area and at least one other geographic area.

In some examples, a particular and wider context may differ for reasons other than the size of the geographic area. For example, a particular context may be associated with a first location, and a wider context may be associated with more than one location, including the first location, at least one other location. Let us consider the example in which each of the several locations is assigned a transport mode of its own. For example, a particular context may be a single airport, and a wider context may include multiple airports. Other examples may be: a single branch and several other branches including the single branch, a residential unit or a home, and several more homes that are not necessarily located in a location, etc.

In some examples, a particular and wider context may differ for reasons other than the size of the location or geographic area. For example, a particular context may include a first amount of time (eg, one hour of a day) and a wider context may include a second amount of time that exceeds the first amount of time (eg, several hours of the day) ,

The need prediction module 364 can table 370 based on contextual information and search queries created over the network 330 from the computer devices 310 were received. The need prediction module 364 receives: query 316B from the computing device 310B while the computer device 310B at the location 304 is positioned, query 316C from the computing device 310C while the computer device 310B at the location 302 is positioned, query 316D from the computing device 310D while the computer device 310D at the location 304 is positioned and query 316E from the computing device 310E while the computer device 310E at the location 304 is positioned. From the table 370 can be the need prediction module 364 Generate a rule that returns a probability of a search performed on a terminal card of the SFO when one of the computing devices 310 in one location 304 (eg within the SFO) and if one of the computer devices 310 within the location 302 located.

In the example of 3 can be a user of the computer device 310A traveling in an airplane that is at the location 304 arrives. While it is at the location 304 the computer device can be located 310A over the network 330 Context information together with the ISS 360 use. The need prediction module 364 can determine that the computer device 310A at the location 302 located and the table 370 Consult to determine if the user of the computer device 310A probably will need some information to help the user perform a predicted task.

In some examples, the fact may be the ISS 360 Information for the computing device 310A to complete a task, depending on whether the user and / or the computing device 310A associated context is new or not. In other words, before providing information, the need prediction module may 364 determine if the current context of the computing device 310A There is a new context that is not in the user of the computing device 310A associated context history is included. If the prediction module 364 determines that the context is not new, then the prediction module 364 refrain from providing information as the predictor module 364 can assume that the user already knows how to complete the task, and to navigate the situation.

In any case, the need prediction module 364 determine that the computer device 310A together with the computing devices 310B . 310C and 310E uses a similar context, each searches for an SFO terminal card performed as the computer device 310A at the location 302 is positioned. Based on the determination that the computer device 310A the same context as the devices 310B . 310C and 310E can share the needs prediction module 364 derive that the user of the computing device 310A can perform a similar action and possibly perform a search for a terminal map of the SFO. Therefore, the needs prediction module 364 instead of the user of the computing device 310A Requiring a search query, as other users have previously done to search for a terminal card, automatically retrieves information 314 provide that represent a graphical image of the terminal card, otherwise the user would probably have to search.

In some examples, the needs prediction module 364 the information from the table 370 aggregate with similar information with other airports or other locations similar to the locations 302 and 304 to determine a more general rule for the information a user may need. Using the general rule, the need prediction module 364 provide different information for completing the task for a different context that excludes the particular context. For example, if the user has landed at the Los Angeles Airport (LAX) airport, the user may still need a terminal card as if he or she had landed on the SFO, but the need prediction module may 364 Instead of providing a graphical image of SFO, logically provide a terminal card from LAX.

In some examples, the needs prediction module 364 provide a service or tool, the third application developer for computing devices 310 enabled the predictor module 364 to inquire about the needs of users based on observed actions. For example, a developer who is interested in content on one of the computing devices 310 to look up the best context to output the financial information by querying the Prediction module 364 for "financial information". The need prediction module 364 can analyze the context histories of devices that searched for RDM symbols, searches that search for results that are related to financial information, device visits 310 on finance-related websites and opening applications related to finance. The need prediction module 364 may provide the developer with the context in which the need for financial information is triggered, in addition to potentially providing different metrics in context so that when and where the need is triggered, such. B. coverage, quality metrics, etc.

4A and 4B are flowcharts, the exemplary operations 410 - 436 Illustrated by an exemplary computer system configured to predict the needs of a user of a computing device, and to provide information to meet the needs, in accordance with one or more aspects of the present disclosure. 4A and 4B are down in the context of system 100 from 1 described. For example, the ISS 160 operations 410 - 436 for predicting the needs of a user of a computing device 110A and provide information to meet the needs in accordance with one or more aspects of the present disclosure.

As in 4A shown, the ISS 160 in operation, receive search histories that correspond to a group of computing devices for a particular context ( 410 ) assigned. For example, users of the computing devices 110 respective query modules 122 prompt queries to the ISS 160 to send searches for information. In addition to performing searches of the search queries, the search engine may 166 Store the search queries and / or results to generate one or more search histories that are common to each of the computing devices 110 assigned.

The ISS 160 can identify, based on the search traces, a task that is performed by users of the group of computing devices for the particular context ( 412 ) is performed. For example, a needs prediction module 164 one or more correlations between the context of computing devices 110 and determine the search histories generated in each of the contexts, and may derive one or more searches that the computer devices 110 if they are in similar contexts.

The ISS 160 may determine a first degree of probability that the task is performed by the users of the group of computing devices for the particular context ( 414 ) is performed. For example, the needs prediction module 164 for each potential task for each given context, determine a set of searches based on the search histories of the group of computing devices 110 associated with each of the tasks. The need prediction module 164 may determine that these tasks, which have a larger amount of searches, have a greater likelihood that it is a task that can be performed by a user for the particular context.

In some examples, the needs prediction module 164 determine the first degree of probability, at least in part, based on a first set of searches (ie, a first number of searches) from the scans, that of the group of computing devices 110 for a particular context associated with the task and a total set of searches (ie, total number of searches) based on the search histories associated with the group of computing devices for the particular context. In other words, the need prediction module 164 Instead of relying on counts per se, they modulate specific counts to rely on percentages or ratios for each task performed compared to all tasks performed in a particular context to determine a degree of probability that a certain task is performed for a given context.

As used herein, the need prediction module 164 derive based on one or more of the machine learning rules to which the need prediction module 164 Having access to predict the needs of users based on actions that a search is "associated with a task" or "associated with a task". The rules may be based on an assumption that the searches that produce the same results or perform similar actions are related to the same needs. For example, when users enter searches for terminal cards, airport floor transportation, baggage area locations, etc. in an engine input, the search engine may produce an airport terminal card as the best result. Therefore, the needs prediction module 164 derive that searches for terminal cards, airport floor transport, locations of luggage areas, etc. are all related to the need for a terminal card.

The ISS 160 may determine a second degree of likelihood that the task will be performed by the users of the group of computing devices for a broader context, the particular context and at least one other context ( 416 ) includes. For example, the needs prediction module 164 For each potential task for broader contexts, determine an additional set of searches based on the search histories of the group of computing devices 110 associated with each of the tasks. The need prediction module 164 can determine that these tasks, which have a larger amount of searches for the broader contexts, also have a greater likelihood of being one Task that can be performed by a user for the wider context.

The ISS 160 can determine if the first degree of probability is the second degree of probability by a threshold amount ( 418 ) exceeds. For example, the needs prediction module 164 instead of providing information to complete a task every time the ISS 160 assume that a person may need information to complete a task, determine whether the need is truly unique to a particular context, or instead, whether the need is well-known in view of the popularity of the need in broader contexts.

In response to determining that the first degree exceeds the second degree by a threshold amount, the ISS module may 160 Determine if the current context of a particular computing device 110 the particular context ( 420 ) associated with the derived task. For example, the needs prediction module 164 if that through the need prediction module 164 The model used is contextual information about the computing devices 110 Enter into the model to determine if the current context is from any of the computing devices 110 is related to or corresponds to a context with an observed need. The prediction module 164 can determine that the computer device 110A has a context that is typically associated with the need for ground transportation information from an airport in another country.

In response to determining that the current context of the particular computing device 110 In accordance with the particular context, the ISS 160 to the particular computing device information about completing the task for the particular context ( 422 ) transfer. For example, the needs prediction module 164 in response to determining that a user of the computing device 110A could be in a situation where the user needs a taxi service, the search module 166 for information about taking taxis from the particular airport in the other country. The need prediction module 164 Can the through the search engine 166 retrieved information as content for presentation to the UID 112A on the computer device 110A package, and content as data over the network 130 to the UI module 120A transfer. The UI module 120A can the UID 112A configure to present the content (eg on a display).

As in 4B shown, the ISS 160 in operation, based on the search histories, identify a second task performed by users of the group of computing devices for the particular context ( 424 ) is performed. For example, the needs prediction module 164 In addition to determining that users search for ground transportation information at the airport in the other country, they also determine that users of computing devices 110 often looking for a terminal map of the airport.

The ISS 160 may determine a third degree of likelihood that the second task will be performed by the users of the group of computing devices for the particular context ( 426 ), and may further determine a fourth degree of likelihood that the second task will be performed by the users of the group of computing devices for the broader context, the particular context, and at least one other context ( 428 ) includes. For example, the needs prediction module 164 similar to the processes described above 414 and 416 determine respective degrees of probability associated with the task of searching for a terminal card in both the context of the airport in the other country and the context of the entire other country.

In some examples, the ISS 160 in response to determining that the third degree of probability exceeds the fourth degree of probability by the threshold amount and that the current context of the particular computing device corresponds to the particular context to which particular computing device transmits second information for completing the second task for the particular context , For example, the needs prediction module 164 just as with the processes described above 418 - 422 in response to determining that a user of the computing device 110A in a situation where the user needs a terminal card, the search engine 166 for information about terminal maps for the particular airport in the other country. The need prediction module 164 Can the through the search engine 166 retrieved information as content for presentation to the UID 112A on the computer device 110A package, and content as data over the network 130 to the UI module 120A transferred, with the UI module 120A the UID 112A configure to present the content (eg on a display).

The ISS 160 can determine a classification of the first and second task and the classification of the computing device 110A deploy it so that the computing device 110A can distinguish what information on the UID 112A to show (eg the terminal card or the taxi information).

To determine the classification, the ISS 160 a first difference between the first degree of probability and the second degree of probability ( 430 ) and determine a second difference between the third degree of probability and the fourth degree of probability ( 432 ). Based on the first difference and the second difference, the ISS 160 determine the classification of the first task and the second task. For example, the ranking of the first and second tasks is not necessarily based on which task has the most counts or the highest probability for a particular context. Instead, the rating may be based on which task has the most counts or the highest probability for a particular context compared to a threshold. The threshold may be representative of the total number of counts or the probability of the wider context. In other words, the ISS 160 The task, which is more unique to the particular situation, is rated higher for the reason that it is less likely that the more unique task is known or understood by a user in a new situation.

For example, the needs prediction module 164 determine a first set of searches from the search histories associated with the group of computing devices for the particular context associated with or otherwise associated with the first task and a second set of searches from the search histories belonging to the group of Associated with the particular context associated with or otherwise associated with the second task. Then the need prediction module 164 which classify two tasks based on the two sets when compared to respective sets of the first and second tasks for broader contexts.

In any case, the ISS 160 after classifying the two tasks, transferring the rating to the computing device to switch between outputting the first information and the second information ( 436 ) to prioritize. For example, the needs prediction module 164 Information about the network 130 which indicate the classification between presenting the terminal card or the taxi information. Upon receipt of the classification, the UI module may 120A the UID 112A to present the higher rated information first.

The ISS 160 can build the categorization of needs and tasks based on the idea that it is possible to search the action logs and count how many times a need or a task occurred for a particular need and context, as actions using action logs (e.g. B. Search curves) are observable. That way, the ISS 160 compare the counts associated with other tasks as a way to compare the needs relative to one another (eg, important to score / score), and in absolute terms (eg, to trigger when the need is to be presented or transmitted) ,

For example, the ISS 160 on a score function s (N, C) used for ranking the candidate needs N as a function f based on the probability P (N | C) that the user has the need N in view of the user context C, as in GL. 2: N * = argmax _N P (N | C) / P (N) GL. 2

Given a given context C, the ISS 160 P (N | C) may be used to predict the largest candidate needs, possibly as an input to a function that modulates the likelihood of a known need versus the "wow" factor of a need of interest (eg, "raising" a need of interest) a known need). That is, the probability given the context divided by the background probability for the need. For example, based on the relationship described in GL. 3 is shown: N * = argmax _N P (N | C) / P (N | C) GL. 3

In some examples, the ISS 160 instead of using probabilities, assess needs based on counts with scores. GL. 4 defines a need score as: Score = # watched - #expected (background) - # noise (# watched) GL. 4

In this derivation # Observed is the number of times an action by the need prediction module 164 in the search histories and / or context information of the computing devices 110 was observed, #expected (background) is the expected number of times that the ISS 160 expects to observe an action in the current context assuming that the action appears similar in the given background (e.g., [# count in background / # all counts in background] × # all counts in context). Finally, Noise (#Watched) is a normalization factor to the ISS 160 under the provision that actions with low counts may not be relevant to the need prediction. In some examples a formulation of # noise (#beobachtet) a smooth function such. B. be an inverse sigmoidal function. A simple formulation for noise (#Watched) could be the one in GL. 5 function shown: # Noise (#Watched) = #Watched if #Watched <10 # Noise (#Watched) = 0, otherwise GL. 5

 Clause 1. A method, comprising: identifying, by a computer system, a task performed by users of the group of computing devices for a particular context based on search histories associated with a group of computing devices for a particular context; Determining, by the computer system, a first degree of likelihood that the task is performed by the users of the group of computing devices for the particular context; Determining by the computer system a second degree of likelihood that the task is performed by the users of the group of computing devices for a wider context including the particular context and at least one other context; and in response to determining that the first degree of probability exceeds the second degree of probability by a threshold amount, and that a current context of a particular computing device corresponds to the particular context that informs the computer system of completing the task for the particular context certain computer device transmits.

Clause 2. The method of clause 1, wherein the information for completing the task for the particular context is further transmitted in response to determining that the context is a new context that is not included in a context history associated with the user of the computing device.

Clause 3. The method of any of clauses 1-2, wherein the information for completing the task for the particular context is first information, the method further comprising: determining, by the computer system, second information different from the first information to complete the task for a different context that excludes the particular context.

Clause 4. The method of any one of clauses 1-2, wherein the first degree of likelihood is determined, at least in part, based on a set of searches from the search histories associated with the group of computing devices for the particular context associated with the task.

Clause 5. The method of clause 4, wherein: the set of searches is a first set of searches of first searches associated with a first group of computing devices for the particular context, and the second degree of probability based at least in part on a second set is determined by searches of search histories associated with a second group of computing devices for the wider context associated with the task.

Clause 6. The method of any one of clauses 1-5, wherein the particular context includes a first amount of time, and the broader context includes a second amount of time greater than the first amount of time.

Clause 7. The method of any of clauses 1-6, wherein the particular context is associated with a first geographic area and the broader context is associated with a second geographic area that includes the first geographic area and at least one other geographic area.

Clause 8. The method of any of clauses 1-7, wherein the information for completing the task for the particular context is based on at least one search result from the search histories associated with the group of computing devices.

Clause 9. A computer system comprising: at least one processor and at least one module operable by the at least one processor to: identify a task performed by users of the group of computing devices for the particular context based on search histories associated with a group of computing devices for a particular context assigned to a specific context; Determining a first degree of likelihood that the task is performed by the users of the group of computing devices for the particular context; Determining a second degree of likelihood that the task is performed by the users of the group of computing devices for a wider context that includes the particular context and at least one other context; and in response to determining that the first degree of probability exceeds the second degree of probability by a threshold amount and that a current context of a particular computing device corresponds to the particular context, transmitting information to complete the task for the particular context particular computing device.

Clause 10. The computer system of clause 9, wherein the particular context is associated with a first geographic area, and the broader context is associated with a second geographic area that includes the first geographic area and at least one other geographic area.

Clause 11. The computer system of any of clauses 9-10, wherein the first degree of likelihood is determined, at least in part, based on a first set of searches from the search histories associated with the group of computing devices for the particular context associated with the task , and a total of searches, from the search histories associated with the group of computing devices for the particular context.

Clause 12. The computer system of any one of clauses 9-11, wherein the particular context is associated with a first site and the broader context is associated with multiple sites including the first site and at least one other site.

Clause 13. The computer system of clause 12, wherein each of the multiple locations is associated with a same transport mode.

Clause 14. The computer system of any of clauses 9-13, wherein the group of computing devices excludes the particular computing device.

Clause 15. The computer system of any one of clauses 9-14, wherein the at least one module is further operable by the at least one processor to refrain from transmitting the information to complete the task for the particular context in response to determining the first degree of probability does not exceed the second degree of probability by the threshold amount.

Clause 16. A computer-readable storage medium comprising instructions that, when executed, configure at least one or more processors of a computer system to: identify a task performed by users of the group of computing devices for a particular context based on search histories associated with a group of computing devices for the particular context are; Determining a first degree of likelihood that the task is performed by the users of the group of computing devices for the particular context; Determining a second degree of likelihood that the task is performed by the users of the group of computing devices for a wider context that includes the particular context and at least one other context; and in response to determining that the first degree of probability exceeds the second degree of probability by a threshold amount and that a current context of a particular computing device corresponds to the particular context, transmitting information to complete the task for the particular context particular computing device.

Clause 17. The computer readable storage medium of clause 16, wherein the task is a first task, the information is first information, and configuring the instructions in executing the one or more processors of the computer system to: determine a third degree of probability that a second one Task is performed by the users of the group of computing devices for the particular context, the second task being further identified based on the lookups associated with the group of computing devices for the particular context; and determining a fourth degree of likelihood that the second task is performed by the users of the group of computing devices for the wider context including the particular context and at least one other context; and in response to determining that the third degree of probability exceeds the fourth degree of probability by the threshold amount and that the current context of the particular computing device corresponds to the particular context, transmitting information to complete the second task for the particular context the particular computing device.

Clause 18. The computer-readable storage medium of clause 17, wherein the instructions, when executed, further configure the one or more processors of the computer system to: determine a first difference between the first degree of probability and the second degree of probability; Determining a second difference between the third degree of probability and the fourth degree of probability; Determining a rating of the first task and the second task based on the first difference and the second difference; and transmitting the rating for prioritizing between outputting the first information and the second information to the computing device.

Clause 19. The computer-readable storage medium of clause 18, wherein the instructions, when executed, further configure the one or more processors of the computer system to determine the ranking of the first and second tasks based on: a first set of searches from the lookups belonging to the group of Computer devices associated with the particular context are assigned to the first task are; and a second set of searches from the search histories associated with the set of computing devices for the particular context associated with the second task.

Clause 20. The computer-readable storage medium of any of clauses 16-19, wherein the task includes navigating to a particular location from a current location of the particular computing device, and the information includes one or more suggestions for traveling from the current location to the particular location.

Clause 21. The computer system of clause 9, comprising means for performing one of the methods of any of claims 1-8.

Clause 22. The computer readable storage medium of clause 16, comprising further instructions that, when executed, cause the one or more processors of the computer system of clause 9 to perform one of the methods of clauses 1-8.

In one or more examples, the described functions may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored or transmitted as one or more instructions or codes as a computer-readable medium and executed by a hardware-based processing unit. The computer-readable medium may include computer-readable storage media or media that may be stored on a physical medium, such as a computer. As data storage media, or communication media, including media correspond to the transfer of a computer program from one location to another, z. B. according to a communication protocol facilitate. In this way, the computer-readable medium may generally correspond to physical computer-readable (1) storage media that is non-volatile or (2) a communication medium, such as a signal or carrier wave. Data storage media may be any available media that may be accessed by one or more computers or one or more processors to retrieve instructions, code, and / or data structures for implementing the techniques described in this disclosure. A computer program product may include a computer readable medium.

By way of example and not limitation, such computer readable storage media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory, or any other media that may be used to provide the desired Program code in the form of instructions or data structures to store and which can be accessed by a computer. In addition, each connection is referred to as a computer-readable medium. For example, when instructions are transmitted from a web page, server, or other remote source using a coaxial cable, a fiber optic cable, a twisted pair cable, a digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave Coaxial cable, fiber optic cable, twisted pair cable, DSL or wireless technologies such as infrared, radio and microwave included in the definition of medium. It should be understood, however, that computer-readable storage media and media, as well as data storage media, do not include interconnections, carrier waves, signals, or other physical media, but instead are targeted to nonvolatile physical storage media. Hard disks and floppy disks as used herein include Compact Disc (CD), Laser Disc, Optical Disc, Digital Versatile Disc (DVD), floppy disk, and Blu-ray Disc, where floppy disks typically magnetically display data, while discs optically reproduce data with lasers , Combinations of the above should also be included within the scope of a computer-readable medium.

Instructions may be executed by one or more processors, such as one or more digital signal processors (DSPs), general purpose microprocessors, application oriented integrated circuits (ASICs), field programmable gate arrays (FPGAs), or any other equivalent integrated or discrete logic circuits. Accordingly, the term "processor" as used herein may refer to any of the foregoing structures or any other structure suitable for implementing the techniques described herein. In addition, in some aspects, the functionality described herein may be provided within dedicated hardware and / or software modules. Also, the techniques could be fully implemented in one or more circuits or logic elements.

The techniques of the present disclosure may be implemented in a wide variety of devices or devices, including a wireless handset, an integrated circuit (IC), or a set of ICs (eg, a chipset). In this disclosure, various components, modules, or devices are described to emphasize functional aspects of devices configured to perform the disclosed techniques, but do not necessarily require implementation by various hardware devices. Rather, as described above, various units may be combined in a hardware unit or provided by a collection of interoperable hardware units, including one or more processors as described above, in conjunction with appropriate software and / or firmware.

 Different embodiments have been described. These and other embodiments are within the scope of the following claims.

Claims (15)

 Method, comprising: identifying, by a computer system, a task performed by users of the group of computing devices for a particular context based on search histories associated with a group of computing devices for a particular context; Determining by the computer system a first degree of likelihood that the task is performed by the users of the group of computing devices for the particular context; Determining by the computer system a second degree of likelihood that the task is performed by the users of the group of computing devices for a wider context including the particular context and at least one other context; and in response to determining that the first degree of probability exceeds the second degree of probability by a threshold amount and that a current context of a particular computing device corresponds to the particular context, transmitting information to complete the task for the particular context to the particular computing device; through the computer system.  The method of claim 1, wherein the information for completing the task for the particular context is further transmitted in response to determining that the context is a new context that is not included in a context history associated with the user of the computing device.  The method of any one of claims 1-2, wherein the information for completing the task for the particular context is first information, the method further comprising: determining second information other than the first information to complete the task for another context that excludes the particular context.  The method of claim 1, wherein the first degree of probability is determined based at least in part on a first set of first searches of first search histories associated with the group of computing devices for the particular context associated with the task, and the second degree of probability is determined based at least in part on a second set of second search history searches associated with a second group of wider context computing devices associated with the task.  The method of any of claims 1-4, wherein at least one of: the particular context comprises a first amount of time, and the wider context includes a second amount of time greater than the first amount of time; or the particular context is associated with a first geographic area and the wider context is associated with a second geographic area that includes the first geographic area and at least one other geographic area.  The method of any of claims 1-5, wherein the information for completing the task for the particular context is based on at least one search result from the search histories associated with the group of computing devices.  Computer system comprising: at least one processor; and at least one module that can be used by the at least one processor for: Identifying a task performed by users of the group of computing devices for the particular context based on search histories associated with a group of computing devices for the particular context; Determining a first degree of likelihood that the task is performed by the users of the group of computing devices for the particular context; Determining a second degree of likelihood that the task is performed by the users of the group of computing devices for a wider context that includes the particular context and at least one other context; and in response to determining that the first degree of probability exceeds the second degree of probability by a threshold amount and that a current context of a particular computing device corresponds to the particular context, transmitting information to complete the task for the particular context to the particular one computing device. The computer system of claim 7, wherein the first degree of probability is determined, at least in part, based on a first set of searches from the search histories associated with the group of computing devices for the particular context associated with the task and a total amount of searches, of the search histories associated with the group of computing devices for the particular context.  The computer system of any one of claims 7-8, wherein the particular context is associated with a first location and the broader context is associated with a plurality of sites including the first site, at least one other site, and wherein each of the plurality of sites is associated with a self-transport mode.  The computer system of any one of claims 7-9, wherein the group of computing devices excludes the particular computing device.  The computer system of claim 7, wherein the at least one module is further operable by the at least one processor to refrain from transmitting the information for completing the task for the particular context in response to determining that the first degree Probability does not exceed the second degree of probability by the threshold amount.  The computer system of any of claims 7-11, wherein the at least one module is further capable of the at least one processor for: Determining a third degree of likelihood that a second task is being performed by the users of the group of computing devices for the particular context, the second task further being identified based on the lookups associated with the group of computing devices for the particular context; and Determining a fourth degree of likelihood that the second task is performed by the users of the group of computing devices for the wider context including the particular context and at least one other context; and in response to determining that the third degree of probability exceeds the fourth degree of probability by the threshold amount, and that the current context of the particular computing device corresponds to the particular context, transmitting second information to complete the second task for the particular context the particular computing device.  The computing device of claim 12, wherein the at least one module further of the at least one processor is capable of: Determining a first difference between the first degree of probability and the second degree of probability; Determining a second difference between the third degree of probability and the fourth degree of probability; Determining a rating of the first task and the second task based on the first difference and the second difference; and Transmitting the rating for prioritizing between outputting the first information and the second information to the computing device.  The computer system of any of claims 7-13, wherein the task includes navigating to a particular location from a current location of the particular computing device, and the information includes one or more suggestions for traveling from the current location to the particular location.  A system comprising means for performing any of the methods of claims 1-6.

Images